OP67The changing food environment and area level deprivation: an observational study of takeaway outlet and supermarket exposure in Norfolk, 1990-2008

BackgroundAlthough previous studies have revealed an association between takeaway prevalence and deprivation, few UK-based studies have investigated associations between deprivation and both healthy and unhealthy outlets or trends in prevalence over time. Given increasing policy interest in food environments it is important to examine long term trends by level of deprivation. This study examined takeaway and supermarket prevalence by area-level deprivation in Norfolk during 1990-2008.MethodsWe recorded food outlet data from Yellow Pages directories over six time points from 1990 to 2008. We geocoded and mapped addresses onto 2001 electoral ward boundaries. We classified 'Supermarkets' and 'Takeaways' using a priori definitions based on market share and planning regulations. We population-standardised outlet counts using the 2001 Census and assigned wards to deprivation tertiles using the 2001 Townsend Index. We used repeated measures ANOVA to test differences in takeaway density over time and multilevel logistic regression to test the interaction between deprivation tertile and time in the presence of supermarkets.ResultsMean takeaway prevalence increased from 2.6 per 10,000 residents in 1990 to 3.7 in 2008. Supermarket prevalence increased from 0.2 to 0.3 per 10,000 residents over this time. Takeaway prevalence was significantly greater in the most deprived tertile throughout the period: in 2008 there were 2.1 (SD=4.3) outlets in the least deprived tertile, 2.7 (SD=5.7) in the middle tertile, and 6.5 (SD=7.6) in the most deprived. Takeaway prevalence increased significantly (F=12.8, p < 0.001), but there was no interaction between deprivation and time (F=1.7, p = 0.11). There was no significant association between deprivation and the odds of supermarket presence and no significant interaction between deprivation tertile and year.ConclusionWe found a greater prevalence of takeaway outlets in more deprived wards than less deprived wards. However, we found no difference in the odds of a supermarket being present. Over time there was an increase in both the prevalence of supermarkets and takeaways but this did not differ by deprivation. This is the first UK study to assess the local food environment in relation to area deprivation over time. Future work needs to assess a broader range of outlets to better capture the food environment and determine the extent to which exposure affects dietary behaviours.